Food Chemistry 174 (2015) 124131

Contents lists available at ScienceDirect

Food Chemistry
journal homepage: www.elsevier.com/locate/foodchem

Effects of avonoid glycosides obtained from a Ginkgo biloba extract

fraction on the physical and oxidative stabilities of oil-in-water
emulsions prepared from a stripped structured lipid
with a low omega-6 to omega-3 ratio
Dan Yang a, Xiang-Yu Wang a, Lu-Jing Gan c, Hua Zhang c,b, Jung-Ah Shin c, Ki-Teak Lee c,
Soon-Taek Hong c,
a

Food Quality & Safety Center, Nutrition & Health Research Institute, COFCO Corporation, Beijing, Changping 102-209, China
Department of Food Science and Engineering, Yanbian University, Jilin Province, Yanbian 133-000, China
c
Department of Food Science and Technology, Chungnam National University, Daejeon 305-764, South Korea
b

a r t i c l e

i n f o

Article history:
Received 3 June 2013
Received in revised form 31 October 2014
Accepted 6 November 2014
Available online 13 November 2014
Keywords:
Structured lipid
Ginkgo biloba
Flavonoid glycosides
Antioxidant capacity
Emulsion stability

a b s t r a c t
In this study, we have produced a structured lipid with a low x6/x3 ratio by lipase-catalysed interesterication with perilla and grape seed oils (1:3, wt/wt). A Ginkgo biloba leaf extract was fractionated in a
column packed with HP-20 resin, producing a avonoid glycoside fraction (FA) and a biavone fraction
(FB). FA exhibited higher antioxidant capacity than FB, showing 58.4 mmol gallic acid equivalent
(GAE)/g-of-total-phenol-content, 58.8 mg quercetin equivalent (QUE)/g-of-total-avonoid-content,
4.5 mmol trolox/g-of-trolox-equivalent antioxidant capacity, 0.14 mg extract/mL-of-free-radical-scavenging-activity (DPPH assay, IC50), and 2.3 mmol Fe2SO47H2O/g-of-ferric-reducing-antioxidant-power.
The oil-in-water emulsion containing the stripped structured lipid as an oil phase with FA exhibited
the highest stability and the lowest oil globule diameters (d43 and d32), where the aggregation was unnoticeable by Turbiscan and particle size analyses during 30 days of storage. Furthermore, FA was effective
in retarding the oxidation of the emulsions.
2014 Elsevier Ltd. All rights reserved.

1. Introduction
Structured lipids (SLs) are triacylglycerols (TAGs) that have
been modied by chemical or enzymatic reactions to change the
fatty acid composition and/or their location in the glycerol backbone (Hamam & Shahidi, 2004). SLs provide the ability to create
custom-made lipids with health benets through esterication.
Perilla oil is one of the major sources of plant omega-3 fatty acids
(C18:3, x-3, ALA). ALA has been proven to prevent cardiovascular
disease, help manage chronic disorders, and decrease blood cholesterol levels (Sanders et al., 1997). However, western diets are usually decient in x-3 fatty acids and have excessive amounts of
omega-6 (x-6) fatty acids (Simopoulos, 2002). It has been claimed
that the intake of fats with a low x6/x3 ratio would help in preventing coronary heart and atherosclerotic diseases (Simopoulos,
2002; Harris et al., 2009), and the use of SLs is an alternative
method for obtaining such benets.
Corresponding author. Tel.: +82 42 821 6727; fax: +82 42 821 8900.
E-mail address: hongst@cnu.ac.kr (S.-T. Hong).
http://dx.doi.org/10.1016/j.foodchem.2014.11.036
0308-8146/ 2014 Elsevier Ltd. All rights reserved.

Flavonoids are a class of plant secondary metabolites, the dietary consumption of which might be related to protection against
some diseases (Hertog, Feskens, Kromhout, Hollman, & Katan,
1993). Furthermore, as an additive, avonoids can retard lipid oxidation in emulsion systems (Roedig-Penman & Gordon, 1998).
Recently, studies have shown that avonoids can act as stabilizers
of oil-in-water (O/W) emulsions through pickering stabilisation
(Luo et al., 2011a, 2011b). This nding indicates that while the
avonoids exist as insoluble particles in the aqueous phase, they
tend to adsorb at the oil-water interface. The adsorbed layer could
provide an efcient steric barrier against the coalescence of emulsion oil globules. Ginkgo biloba leaf ethanol extracts (GBEs) have
been widely investigated because of their possible benecial
effects on human health, such as alleviating short-term memory
loss and disturbances in vigilance and mental concentration
(Kobus et al., 2009). Furthermore, several studies have shown the
antioxidant activity of GBEs in vitro and in vivo (Kobus et al.,
2009). Since avonoids are claimed to be the main bioactive compounds in GBEs, it is important to know if the fractionated extracts,
which mainly contain avonoid glycosides and biavones, still

D. Yang et al. / Food Chemistry 174 (2015) 124131

retain their bioactivity after being incorporated in the emulsions

and how they may interfere with the physical and chemical stabilities of the emulsions. Specically, recent studies (Atars, Marshall,
Akhtar, & Murray, 2012; Luo et al., 2011a, 2011b) have found that
some avonoids show good emulsifying activity, indicating that
the GBE particles may help maintain the physical and chemical stabilities of O/W emulsions.
To date, only a few methods have been developed for monitoring the destabilisation process of emulsions, such as optical analysis with microscopy, spectroscopy, turbidity analysis, and particle
size analysis (McClements, 2007). However, these methods involve
some form of dilution, which is not suitable for investigating the
destabilisation process. Turbiscan optical analyzer can be used
for real-time monitoring of the turbidity prole of an emulsion
along the height of a glass tube lled with the emulsion (Liu
et al., 2011; Mengual, Meunier, Cayre, Puech, & Snabre, 1999;
Pan, Toms, & An, 2004; lvarez Cerimedo, Iriart, Candal, &
Herrera, 2010). It provides a method for quick and simple detection
of changes in the physical stability of the emulsion (creaming, sedimentation, clarication, coalescence, and occulation), by recording the transmission and backscattering variations in the emulsion
over time (Palazolo, Sorgentini, & Wagner, 2005; Huck-Iriart,
Pizones Ruiz-Henestrosa, Candal, & Herrera, 2013).
Our previous study (Yang et al., 2013) showed that 80% GBE,
containing both avonoids and biavones, could act as a good antioxidant in an O/W emulsion prepared from SLs and physically
blended lipids. In the current study, we obtained samples with different GBE fractions containing either avonoid glycosides or biflavones, and we examined these fractions for their antioxidant
activity with regards to inhibiting lipid oxidation in the O/W emulsions prepared from stripped structured lipids (SSLs). Finally, the
physical and oxidative stabilities of the O/W emulsions were monitored during 30 days of storage.
2. Materials and methods
2.1. Materials
Grape seed oil (GSO) and perilla oil (PO) were purchased from
the local market (Daejeon, South Korea). The GSO contained 6.6%
C16:0, 4.8% C18:0, 17.4% C18:1, 70.9% C18:2 (x-6), and 0.2%
C18:3 (x-3) fatty acids, whereas the PO contained 6.3% C16:0,
1.9% C18:0, 21.0% C18:1, 10.4% C18:2 (x-6), and 60.5% C18:3 (x3). The SL contained 6.0% C16:0, 3.3% C18:0, 17.3% C18:1, 57.6%
C18:2 (x-6), and 15.8% C18:3 (x-3), with an x6 to x3 ratio of
3.6. G. biloba leaf was locally collected in early October. TBA (4,6dihydroxy-2-mercaptopyrimidine),
trolox
(6-hydroxy-2,5,7,
8-tetramethylchroman-2-carboxylic acid), ABTS (2,20 -azinobis-3ethylbenzothiazoline-6-sulphonic acid), TPTZ (2,4,6-tripyridyl-striazine), ferrous chloride, sodium acetate trihydrate, gallic acid,
FolinCiocalteus phenol, 2,2 diphenyl-1-picryihydrazyl, quercetin
(Que),
bis[2-hydroxyethyl]amino-tris-[hydroxymethyl]methane
(bis-tris), Diaion HP-20, alumina, silicic acid, charcoal, lecithin,
and chloroform-d (99.9 atom% D, containing 0.1% v/v TMS) were
purchased from SigmaAldrich (St. Louis, MO, USA). All the chemicals and solvents were HPLC-grade and were obtained from Fisher
Scientic (Norcross, GA, USA). Lipozyme TLIM was purchased from
Novozyme A/S (Bagsvaerd, Denmark).
2.2. Production and purication of SLs
The SLs were produced in a 250 mL ask with a screw cap using
the optimal reaction conditions for GSO and PO (3:1, w/w, 24 g).
Lipozyme TLIM (10% of the total substrate) was used for the reaction at 55 C for 6 h (Yang et al., 2013). The tocopherols, b-carotene,

125

diacylglycerol (DAG), monoglycerols (MAG), and free fatty acids

(FFA) were removed from the SLs with some modications (Boon
et al., 2008). The SSLs were prepared by diluting 20 g of SLs with
10 mL hexane. This mixture was passed through a chromatographic column (3.5 cm in diameter and 20 cm in length). The bottom layer of the column was packed with 4 g of charcoal, and
middle and top layers were lled with 20 g aluminium oxide and
20 g silicic acid, respectively. The SSLs were eluted with 200 mL
hexane, and the hexane was subsequently removed under vacuum
using a rotary evaporator (RE 111, Bchi, Flawil, Switzerland) at
38 C. Traces of hexane were further removed by ushing with
nitrogen, and the SSLs obtained were stored at 80 C until use.
The SLs and SSLs were analysed by HPLC, TLC, and 1H NMR to conrm the removal of tocopherols, b-carotene, DAG, MAG and FFA
(Yang et al., 2012).
2.3. DAG, MAG, and TAG analysis by 1H NMR
SLs or SSLs (50 mg) were dissolved in 700 lL CDCl3 and then
placed in an NMR tube (5 mm in diameter) for further analysis.
1
H NMR spectroscopy were performed on a Bruker Avance III 600
spectrometer operated at 600.13 MHz. All the spectra were processed using ACDlabs NMR Processor, version 10.0. The chemical
shifts (d) are reported with reference to tetramethylsilane (TMS)
at d = 0 ppm.
2.4. Preparation of G. biloba extracts and fraction
G. biloba leaves (100 g) were reuxed with 95% ethanol twice,
followed by 70% ethanol twice [sample/solvent (w/v) = 1/6; 2 h
each time]. The combined extract was concentrated using a rotary
vacuum evaporator, and dried in a vacuum freeze dryer to prepare
the nal extract (GBE, 15 g). A part of the GBE (2 g) was fractionated using a column packed with HP-20 (50 g), and eluted with a
gradient of polar solvents [distilled water to ethanol (100:0,
80:20, 50:50, 30:70, to 0:100)] to obtain ve fractions (F1-F5).
The distilled water fraction (F1) containing the pigments was discarded. The other four collected fractions were then combined to
yield two fractions [FA (F2-F3) and FB (F4-F5)], according to their
HPLC patterns, and concentrated using a rotary vacuum evaporator. After drying in a vacuum freezing dryer, FA (450 mg, yield
22.5%) and FB (90 mg, yield 4.5%) were prepared. The compounds
in FA and FB were identied by HPLC-UV, according to the retention time as described in our previous study (Yang et al., 2013).
The extracts were stored in a desiccator at room temperature until
use.
2.5. Determination of antioxidant activity
In order to evaluate the antioxidative activity of GBE, the total
phenol content (TPC), free radical scavenging activity (DPPH assay,
IC50), trolox equivalent antioxidant capacity (TEAC), and ferric
reducing antioxidant power (FRAP) were measured, according to
the methods described by Yang et al. (2013). The total avonoid
content (TFC) was measured, as described by Abdel-Hameed
(2009).
2.6. Preparation of emulsion
Oil-in-water emulsion (100 g) was prepared with 10% (wt.%)
SSL in a 20 mM bistris buffer solution (pH 7). Lecithin was used
as an emulsier at a concentration of 3 wt.% of the SSLs. FA or quercetin was added at a nal concentration of 100 ppm in the mixture.
The chelation of ferric ions by FA or quercetin was also investigated. Ferric chloride (50 lM) was added into the mixture. The
mixture was kept in warm water until pre-homogenised by a

126

D. Yang et al. / Food Chemistry 174 (2015) 124131

Silverson mixer (Model L4RT, Silverson Machines, UK) at a speed of

5000 g for 2 min, and the premix was then passed through a microuidizer (M-110Y, Microuidics, MA, USA) with two passes at
3000 psi. The prepared emulsion was placed in the 50 mL vial with
caps and stored in an oven at 25 C for the oxidative stability study.

The particle size distribution of the oil globules in emulsion was

measured using a laser diffraction instrument (Mastersizer S, Malvern Instrument, Worcestershire, UK) and the results were
expressed as the volume-surface mean diameter (d32) and the
weight mean diameter (d43) (Luo et al., 2011a). The oil globular
sizes of the emulsion samples were recorded at 0 day, 8 day,
16 day, 24 day, and 30 day, after emulsion formation.
The emulsion stability was determined using Turbiscan
(Turbiscan LAB, Formulaction, Toulouse, France) as described elsewhere (Mengual et al., 1999; Pan, Toms, & An, 2002; Pan et al.,
2004; lvarez Cerimedo et al., 2010). One milligram of FA was dispersed in 1 g SSLs with 30 mg lecithin, to which 9 mL of 20 mM
bistris buffer solution (pH 7) was added. The emulsion was
prepared using an ultrasonic processor (Sonics & Materials Inc.,
Newtown, CT), with a frequency of 20 KHz and a power of 750 W
for 20 s. The Turbiscan scans the emulsion sample from the top
to the bottom with near infrared light (k = 880) and measures the
percentage of backscattering (BS) and transmission (T) through
the sample as a function of the height of the tube, in order to quantify the rate of emulsion destabilisation. The backscattering was
measured in 5 min intervals at 22.5 C. Creaming, clarication,
and coalescence/occulation kinetics were followed by determining the mean value kinetics (DBS%) as a function of time at the
clarication layer, middle layer, and creaming layer.
2.8. Emulsion stability (ES) measurements
The ES of the emulsion samples were evaluated using the
method reported by Lima and Alegre (2009), with some modications. The emulsions for ES measurements were prepared in the
same manner as the samples for the Turbiscan measurements
described above. The emulsion (20 lL) was diluted with 8 mL of
bistris buffer. The emulsication activity (EA) was determined
from the absorbance of the emulsion at 540 nm. Each sample
was centrifuged at 3000 rpm for 15 min and the absorbance was
measured at 540 nm. The absorbance was measured at 15 min
intervals for 60 min and the sample was taken from exactly the
same height of the sample tube every time. ES was calculated using
the following equation.

EA
ln
EA0
ES 


kd  t

1
kd

2.10. Statistical analyses

Analysis of variance (ANOVA) was performed by using SPSS 16.0
(SPSS Inc., Chicago, IL). A difference was considered to be statistically signicant when P was less than 0.05.

2.7. Measurement of particle size and Turbiscan analysis

24 day, and 30 days of storage. LOOH and TBARS were determined

according to Yang et al. (2013).

where EA0 in Eq. (1) was the absorbance at the initial time (t = 0). ES
was determined from the slope of the plot of ln (EA/EA0) versus t.
Since the stability of the emulsion was greater for smaller kd values,
the ES value could be used after adjusting by Eq. (2). Quadruplicate
analyses were performed.
2.9. Oxidative stability
Each emulsion (25 mL) prepared using the procedure described
in Section 2.6 was placed in 50 mL vials, and oxidised at 25 C. The
peroxide values (LOOH) and thiobarbituric acid reactive substances
(TBARS) were measured in the emulsion after 0 day, 8 day, 16 day,

3. Results and discussion

3.1. Analysis of neutral lipids in SLs and SSLs
The ALA-enriched SLs were prepared from perilla and grape
seed oils through a lipase-catalysed reaction, where the hydrolysis
and esterication reactions of the TAG molecules occur simultaneously (Willis & Marangoni, 1999). Such a reaction results in
the generation of FFAs, MAGs, and DAGs as by-products, which
would affect the physical and chemical stabilities of the emulsions
prepared from the SLs. Competition between these species and the
emulsiers for adsorption on the oil globule surface is also
expected. Therefore, the FFAs, MAGs, and DAGs were removed
through column chromatography to prepare the SSLs. The amounts
of TAGs, DAGs, and MAGs in the SLs and SSLs were determined by
1
H NMR spectroscopy (Fig. 1). The 1H NMR spectra of the SLs and
SSLs contained multiplets downeld of 5.3 ppm and upeld of
3.8 ppm that were attributed to the b-protons of the acylglyceride,
based on previous literature reports (Laszlo, Compton, &
Vermillion, 2008). After purication through column chromatography, the signals (b-protons) for the DAGs (4.07 ppm and 5.08 ppm
for 1,3-DAG and 1,2-DAG, respectively) and MAGs (4.95 ppm and
3.9 ppm for 2-MAG and 1-MAG, respectively) were not observed
on the 1H NMR spectra corresponding to the SSLs (Fig. 1A). FFAs
were not observed by TLC either (data not shown). However,
before purication, the TAGs, 1,2-DAG, and 1,3-DAG were
82.2 mol%, 8.0 mol%, and 9.8 mol%, of the content of the SLs,
respectively (Fig. 1B). Thus, the DAGs and MAGs in the SSLs were
almost completely removed by the chromatography column
packed with silicic acid, aluminium oxide, and charcoal. Furthermore, the HPLC results showed that the vitamin E and b-carotene
in the SSLs were also removed (data not shown). If these compounds were not removed, they would affect the oxidation stability of the emulsion system.
3.2. Antioxidant capacities of GBE fractions
In our previous study (Yang et al., 2013), we showed that FA
mainly contained avonoid glycosides (e.g., kaempferol, isorhamnetin, apigenin, and quercetin derivatives), whereas FB contained
biavones (e.g., amentoavone, bilobetin, bilobetin isomer,
sequoiaavone, ginkgetin, isoginkgetin, and sciadopitysin). In this
study, the antioxidant capacities of FA and FB were evaluated by
studying the TFC, TPC, DPPH, TEAC, and FRAP contents (Table 1).
The TPC of FA and FB were 58.4 mmol and 66.2 mmol of gallic acid
equivalent (GAE)/g of extract, respectively. In addition, FA showed
a 40% higher TFC, higher antioxidant activity assessed by the DPPH
method with a 10.6 times lower -IC50-, 75% higher TEAC, and 11.5
times higher FRAP compared to the corresponding values obtained
for FB. GBEs have been proven to exhibit strong antioxidant capacities (Goh & Barlow, 2002; Zahradnkov, Schmidt, Sekretr, &
Janc, 2007). Few studies have reported on the antioxidant activities of the fractions obtained from GBEs, although some fractions
were extracted by acetone (Kobus et al., 2009). In this study, the
chromatography column packed with HP-20 resin allowed for
the isolation of avonoid glycosides and biavones from the GBEs.

127

D. Yang et al. / Food Chemistry 174 (2015) 124131

A
100 mol%

B
82.2 mol%

8.0 mol%

9.8 mol%

Fig. 1. Chromatogram of monoacylglycerol (MAG), diacylglycerol (DAG), and triacylglycerol (TAG) in the structured lipids (SL) and stripped structured lipids (SSL) determined
by the 1H NMR (CDCl3, 600 MHz) spectrum of the glycerol backbone region of the SSLs (A) and SLs (B). The peaks assigned to the protons of each acylglycerol species are
labelled.

Table 1
Antioxidant activities of G. biloba fraction.

FA
FB

TPC mmol GAEa/g

extract

Total avonoids
content mg QUEb/g
extract

DPPH
(IC50, mg extract/ml)

TEAC mmol
trolox/g extract

FRAP mmol Fe2SO47H2O/g

extract

58.4 0.01
66.2 0.02

58.8 0.01
41.4 0.01

0.14 0.0
1.63 0.2

4.5 0.01
1.2 0.01

2.3 0.01
0.2 0.01

All data are mean values standard deviations of duplicate measurements.

a
GAE: gallic acid equivalent.
b
QUE: quercetin equivalent.

3.3. Particle size distribution

The oil globule size in the emulsion containing either FA or
quercetin is presented in terms of d32 and d43 (Fig. 2A and B). In
the emulsion with FA, there were almost no changes in d43 and
d32, and this emulsion showed the highest stability among the prepared emulsions (control emulsion and the emulsion with quercetin) during 30 days of storage. However, there was a gradual
increase in d43 and d32 in the control emulsion and the emulsion
with quercetin after 16 days of storage. At 30 days of storage, values of 1.5 lm and 0.7 lm for d43 and d32, respectively, were
observed in the case of the control emulsion, whereas values of

1.4 lm and 0.7 lm were obtained for d43 and d32, respectively,
for the emulsion with quercetin. In the case of the emulsion with
FA, d43 and d32 values of 0.5 lm and 0.3 lm, respectively, were
observed (P < 0.05), indicating that there were barely any changes
in the oil globule size.
Iron in food emulsions has been reported to be a major pro-oxidant that catalyses lipid oxidation (Wang & Wang, 2008). For all
the emulsions containing ferric chloride, the oil globule sizes (d43
and d32) were much higher than those without ferric chloride, suggesting that the iron induces instability of the emulsions prepared
with SSLs (Fig. 2B and D). In addition, the d43 and d32 values in the
control and in the emulsion with quercetin steadily increased after

128

D. Yang et al. / Food Chemistry 174 (2015) 124131

quercetin (1.9 lm for d43 and 0.6 lm for d32) at 30 days of storage,
indicating the aggregation of oil globules in the emulsions. However, in the emulsion containing FA, the d43 and d32 values
remained at 0.7 lm and 0.4 lm, respectively, during the 30 days
of storage, indicating that the FA from the GBE could increase the
stability of the O/W emulsions used in this study.
3.4. Turbiscan analysis

Fig. 2. The weight mean diameter (d43) and the volume-surface mean diameter (d32)
of the emulsion samples during storage. (A) d43 of the emulsion samples, (B) d32 of the
emulsion samples, (C) d43 of the emulsion samples with ferric chloride (50 lM), and
(D) d32 of the emulsion samples with ferric chloride (50 lM). Stripped structured
lipid (SSL) emulsions, control; SSL emulsion with (100 lg/g) of FA, FA; SSL emulsion
with (100 lg/g) of quercetin, Que; SSL emulsions with ferric chloride (50 lM),
control + Fe; SSL emulsion with (100 lg/g) of FA and ferric chloride (50 lM), FA + Fe;
SSL emulsion with (100 lg/g) of quercetin and ferric chloride (50 lM), Que + Fe. Data
shown are averages of duplicate samples. Error bars on the chart represent the range
of the measurements. Different letters denote signicant differences (P < 0.05).

8 days of storage. Fig. 2B and D show that there was a distinct

increase in the mean oil globule sizes (d43 and d32) of the control
(1.7 lm for d43 and 0.8 lm for d32) and the emulsion with

Emulsions are thermodynamically unstable systems, showing

some unstable characteristics such as creaming, occulation, and
coalescence (McClements, 2007). By monitoring the optical properties of an emulsion using the Turbiscan instrument, it is possible
to distinguish between the oil globule occulation and particle
migration processes and obtain real-time information on the
destabilisation process. When the destabilisation process involves
particle migration, a plot of DBS% versus sample height shows a
positive region corresponding to the creamed phase and a negative
lower region caused by clarication (Pan et al., 2004). Fig. 3A
shows the DBS% proles in the reference mode (DBS = BSt  BS0)
as a function of the sample height (total height = 23 mm) for the
emulsions of the control. It is possible to observe three phenomena
simultaneously. First, a clarication process (as evidenced by a
negative DBS% value) takes place at the bottom of the tube (zone
I). Second, an oil globule aggregation process (as evidenced by a
positive DBS% value) takes place at the middle of the tube (zone
II). Third, a creaming process (as evidenced by a positive DBS%
value) takes place at the top of the tube (zone III). BS% is a parameter that is directly dependent on the particles mean diameter (D)
and the particle volume fraction (U), i.e., BS = f (D, U) (lvarez
Cerimedo et al., 2010). According to the BS value obtained in this
study, it was expected that all the emulsions would be homogeneous at t = 0 because the initial mean values of BS% (BS0; i.e., BS
when t = 0) in the control and FA and quercetin-containing emulsions were 84.71%, 85.32%, and 85.86%, respectively (data not
shown). Fig. 3B and D show the DBS% proles (i.e., mean value
kinetics) as a function of time for zones I (clarication, Fig. 3B), II
(oil globule aggregation, Fig. 3C), and III (creaming, Fig. 3D) of
the emulsions. A decrease in DBS% was observed at the bottom
of the tube along with a concomitant increase in DBS% in the upper
zone, owing to the formation of a creaming layer (Huck-Iriart et al.,
2013). However, the degree of clarication was still low because
the clarication phase was still optically opaque and no light
reached the transmission detector (transmission (T%)  0) (Liu
et al., 2011). The slopes for the control and the FA and quercetincontaining emulsions were 0.78, 0.79, and 0.72, respectively
(zone I, Fig. 3B), indicating that there were no signicant differences among the emulsions at the clarication layer.
A comparison of the evolution of the particle size variation
kinetics (zone II) is presented in Fig. 3C. During the 0.4 h analysis,
the DBS% increased to 0.78% and 0.40% for the control and the
emulsion with quercetin, respectively, suggesting an increase in
the oil globule size. In contrast, the emulsion containing FA showed
no distinct increase in the DBS%, which was consistent with the
smaller oil globule size observed in Fig. 2. The increase in DBS%
at the top of the tube was a consequence of the migration of the
ocs (Fig. 3D, zone III). Therefore, FA could slow down the creaming and particle size variation processes of the emulsion system.
The ES values of all the emulsions are presented in Fig. 4. Although
the ES values were not signicantly different (P > 0.05), the emulsion containing FA always showed a higher ES value than those
of the control and the emulsion with quercetin. This ES result indicated that FA possibly improves the ES. It may be noted that the
experiments were performed in quadruplicate.
Based on the above results, the emulsion with FA was proven to
be the most stable among the three kinds of emulsions prepared,

129

D. Yang et al. / Food Chemistry 174 (2015) 124131

90

80

70
Particle variation
Zone II
(5.11mm-16.8mmm)

60
ES (min)

Clarification
Zone I
(0.46mm-5.07mmm)

50
40
30

Creaming
Zone III
(18.7mm-23.3mmm)

20
10
0
Control

0
0.5

1.5

3.5

Slope

-0.5

BS(%)

2.5

Que

Fig. 4. Emulsion stability of the emulsion samples. Stripped structured lipid (SSL)
emulsions, control; SSL emulsion with (100 lg/g) FA, FA; SSL emulsion with
(100 lg/g) quercetin, Que. Data shown are averages of quadruplicate samples. Error
bars on the chart represent standard deviations. Different letters denote signicant
differences (P < 0.05).

B 0.5
0

FA

Control
Que
FA

-1

-0.78
-0.72
-0.79

Control
Que

-1.5

FA

-2
-2.5
-3

Time (h)

C 0.8
0.7

avonoid glycosides. Therefore, such adsorption of avonoid glycosides in the FA can be expected to take place. Once avonoid glycosides are adsorbed at the O/W interface, it is difcult to
remove them from the surface of the oil globule, making the emulsion stable against particle variations (i.e., occulation and coalescence) and migration (i.e., clarication and creaming). Similar
results have been obtained by other researchers (Atars et al.,
2012; Luo et al., 2011a). Luo et al. (2011a, 2011b) reported that
the partition coefcients of avonoids somewhat inuence the surface activity of the avonoids. FA, with its high avonoid glycoside
content, might have suitable partition coefcients, thereby inducing good emulsion stability.

0.6

3.5. Oxidative stability

BS(%)

0.5
0.4

Control

0.3

Que

0.2

FA

0.1
0
0

BS(%)

0.5

1.5
2
Time (h)

2.5

3.5

5
4.5
4
3.5
3
2.5
2
1.5
1
0.5
0

Control
Que
FA

0.5

1.5

2.5

3.5

Time (h)
Fig. 3. Changes in back scattering (BS) proles, (A) DBS% proles as a function of
the sample height corresponding to stripped structured lipid (SSL) emulsions,
control, (B) DBS% proles as a function of time in zone I (clarication), (C) DBS%
proles as a function of time in Zone II (oil globule occulation), (D) DBS(%) proles
as a function of time in zone III (creaming) (mean value) of the emulsion samples,
DBS = BSt  BS0, BSt: backscattering intensity at time = t, BS0: backscattering
intensity at time = 0.

indicating that the presence of FA did exert some effect on the ES.
Luo et al. (2011a, 2011b) reported that avonoids can act as excellent stabilisers of O/W emulsions through their adsorption at the
surface of the oil globule (i.e., pickering stabilisation) as waterinsoluble particles. In this study, FA was found to contain mainly

Lipid oxidation and the generation of secondary oxidation products have always been serious problems in the case of O/W emulsions with respect to food quality and consumer health
(McClements & Decker, 2000). For the control sample, the amount
of LOOHs reached 1.8 mg H2O2/mL of emulsion at 30 days of storage (Fig. 5A). For the emulsions with FA or quercetin, the amount of
LOOHs was signicantly lower than that of the control, with values
of 0.3 mg H2O2/mL and 0.1 mg H2O2/mL, respectively. A similar
trend was observed for the TBARSs (Fig. 5B). The TBARS value for
the control (68 mg TBA/mL of emulsion) was signicantly higher
than that for the emulsion with FA (9.2 mg TBA/mL of emulsion)
or quercetin (1.4 mg TBA/mL of emulsion) during 30 days of storage (P < 0.05).
Metal ions are an important factor that affects lipid oxidation
rates in O/W emulsions (Castellani, Belhomme, David-Briand,
Gurin-Dubiard, & Anton, 2008; Wang & Wang, 2008). Addition
of 50 lM ferric chloride did not signicantly change the overall
trend of LOOH generation in all the emulsion samples (Fig. 5C
and D). The control emulsion also showed the highest amount of
LOOH (2.4 mg H2O2/mL of emulsion), followed by the emulsion
with FA (1.5 mg H2O2/mL of emulsion) and the emulsion with
quercetin (0.27 mg H2O2/mL of emulsion) (P < 0.05). However,
the addition of 50 lM ferric chloride showed a signicant effect
on the generation of TBARSs for the emulsion with FA (Fig. 5D)
compared to the FA-containing emulsion without ferric chloride
(Fig. 5B). The TBARS amount for the emulsion with FA was lower
than that of the control emulsion during 16 days of storage
(P < 0.05), showing 20.3 mg TBA/mL of sample on day 16. Thereafter, it quickly increased to 36.1 mg TBA/mL of sample and 45.6 mg
TBA/mL of sample on days 24 and 30, respectively. However, in
comparison to the emulsion with quercetin, FA does not show a
signicant capability for lowering the primary and secondary

130

D. Yang et al. / Food Chemistry 174 (2015) 124131

FA

2.5

Que

1.5
1
0.5
0

c
0

12
18
Days

c
24

b
b
30

mg H2O2/mL sample

C
Control + Fe

FA+Fe

Que+Fe

a
a
b

a
b

24

30

0
0

12
18
Days

Control

50

FA

Que

30
10

-10

12

b
0

b
18

24

30

Days

70

mg TBA/mL sample

Control

80

mg TBA /mL sample

mg H2O2/mL sample

3.5

Control + Fe

FA+Fe

60

Que+Fe

40

20

a
a

a
b

0
0

12
18
Days

24

30

Fig. 5. Lipid hydroperoxides (LOOH) and thiobarbituric acid-reactive substances (TBARS) of the emulsion samples during the storage. (A) LOOH of the emulsion samples; (B)
TBARS of the emulsion samples; (C) LOOH of the emulsion samples with ferric chloride (50 lM); (D) TBARS of the emulsion samples with ferric chloride (50 lM). Stripped
structured lipid (SSL) emulsions, control; SSL emulsion with (100 lg/g) of FA, FA; SSL emulsion with (100 lg/g) of quercetin, Que; SSL emulsions with ferric chloride (50 lM),
control + Fe; SSL emulsion with (100 lg/g) of FA and ferric chloride (50 lM), FA + Fe; SSL emulsion with (100 lg/g) of quercetin and ferric chloride (50 lM), Que + Fe. Data
shown are averages of duplicate samples. Error bars on the chart represent the range of the measurements. Different letters at the same time denote signicant differences
(P < 0.05).

oxidation products in the emulsion systems with 50 lM of ferric

chloride during 30 days of storage.

4. Conclusions
The properties of O/W emulsions formulated with a low content
of x6/x3 structured lipids and lecithin were signicantly affected
by one of the G. biloba leaf fractions. The emulsions stabilised by
lecithin with FA appeared to be particularly stable with respect
to aggregation, as evidenced by Turbiscan analysis and the smaller
particle size. Oxidation was lower in the FA-containing emulsions
in general, even though oxidation was induced in these emulsions
when Fe3+ was added. Therefore, the results of this study suggest
that GBEs might be able to control the physical and chemical stabilities of structured lipid emulsions containing a high amount of
polyunsaturated fatty acids.
Acknowledgement
This work was supported by the Nuclear Research Development
Project from the Korean Ministry of Science, ICT and Future Planning (Grant No. 2012M2A2A601135).
References
Abdel-Hameed, E. S. S. (2009). Total phenolic contents and free radical scavenging
activity of certain Egyptian Ficus species leaf samples. Food Chemistry, 114(4),
12711277.
lvarez Cerimedo, M. S., Iriart, C. H., Candal, R. J., & Herrera, M. L. (2010). Stability of
emulsions formulated with high concentrations of sodium caseinate and
trehalose. Food Research International, 43(5), 14821493.
Atars, L., Marshall, L., Akhtar, M., & Murray, B. S. (2012). Structure and oxidative
stability of oil in water emulsions as affected by rutin and homogenization
procedure. Food Chemistry, 134(3), 14181424.

Boon, C. S., Xu, Z., Yue, X., McClements, D. J., Weiss, J., & Decker, E. A. (2008). Factors
affecting lycopene oxidation in oil-in-water emulsions. Journal of Agricultural
and Food Chemistry, 56(4), 14081414.
Castellani, O., Belhomme, C., David-Briand, E., Gurin-Dubiard, C., & Anton, M. (2008).
The role of metal ions in emulsion characteristics and occulation behaviour of
phosvitin-stabilised emulsions. Food Hydrocolloids, 22(7), 12431253.
Goh, L. M., & Barlow, P. J. (2002). Antioxidant capacity in Ginkgo biloba. Food
Research International, 35(9), 815820.
Hamam, F., & Shahidi, F. (2004). Synthesis of structured lipids via acidolysis of
docosahexaenoic acid single cell oil (DHASCO) with capric acid. Journal of
Agricultural and Food Chemistry, 52(10), 29002906.
Harris, W. S., Mozaffarian, D., Rimm, E., Kris-Etherton, P., Rudel, L. L., Appel, L. J.,
et al. (2009). Omega-6 fatty acids and risk for cardiovascular disease. Circulation,
119(6), 902907.
Hertog, M. G. L., Feskens, E. J. M., Kromhout, D., Hollman, P., & Katan, M. (1993).
Dietary antioxidant avonoids and risk of coronary heart disease: The Zutphen
elderly study. The Lancet, 342(8878), 10071011.
Huck-Iriart, C., Pizones Ruiz-Henestrosa, V. M., Candal, R. J., & Herrera, M. L. (2013).
Effect of aqueous phase composition on stability of sodium caseinate/sunower
oil emulsions. Food and Bioprocess Technology, 6, 24062418.
Kobus, J., Flaczyk, E., Siger, A., Nogala-Kaucka, M., Korczak, J., & Pegg, R. B. (2009).
Phenolic compounds and antioxidant activity of extracts of Ginkgo leaves.
European Journal of Lipid Science and Technology, 111(11), 11501160.
Laszlo, J. A., Compton, D. L., & Vermillion, K. E. (2008). Acyl migration kinetics of
vegetable oil 1,2-diacylglycerols. Journal of the American Oil Chemists Society,
85(4), 307312.
Lima, . S., & Alegre, R. M. (2009). Evaluation of emulsier stability of biosurfactant
produced by Saccharomyces lipolytica CCT-0913. Brazilian Archives of Biology and
Technology, 52(2), 285290.
Liu, J., Huang, X., Lu, L., Li, M., Xu, J., & Deng, H. (2011). Turbiscan Lab Expert
analysis of the biological demulsication of a water-in-oil emulsion by two
biodemulsiers. Journal of Hazardous Materials, 190(1), 214221.
Luo, Z., Murray, B. S., Ross, A. L., Povey, M. J. W., Morgan, M. R. A., & Day, A. J. (2011a).
Effects of pH on the ability of avonoids to act as Pickering emulsion stabilizers.
Colloids and Surfaces B: Biointerfaces, 92, 8490.
Luo, Z., Murray, B. S., Yusoff, A., Morgan, M. R. A., Povey, M. J. W., & Day, A. J. (2011b).
Particle-stabilizing effects of avonoids at the oil water interface. Journal of
Agricultural and Food Chemistry, 59(6), 26362645.
McClements, D. J. (2007). Critical review of techniques and methodologies for
characterization of emulsion stability. Critical Reviews in Food Science Nutrition,
47(7), 611649.
McClements, D. J., & Decker, E. A. (2000). Lipid oxidation in oil-in-water emulsions:
Impact of molecular environment on chemical reactions in heterogeneous food
systems. Journal of Food Science, 65(8), 12701282.

D. Yang et al. / Food Chemistry 174 (2015) 124131

Mengual, O., Meunier, G., Cayre, I., Puech, K., & Snabre, P. (1999). Characterisation of
instability of concentrated dispersions by a new optical analyser: The
TURBISCAN MA 1000. Colloids and Surfaces A: Physicochemical and Engineering
Aspects, 152(1), 111123.
Palazolo, G. G., Sorgentini, D. A., & Wagner, J. R. (2005). Coalescence and occulation
in o/w emulsions of native and denatured whey soy proteins in comparison
with soy protein isolates. Food Hydrocolloids, 19(3), 595604.
Pan, L., Toms, M., & An, M. (2002). Effect of sunower lecithins on the stability of
water-in-oil and oil-in-water emulsions. Journal of Surfactants and Detergents,
5(2), 135143.
Pan, L., Toms, M., & An, M. (2004). Oil-in-water emulsions formulated with
sunower lecithins: Vesicle formation and stability. Journal of the American Oil
Chemists Society, 81(3), 241244.
Roedig-Penman, A., & Gordon, M. (1998). Antioxidant properties of myricetin and
quercetin in oil and emulsions. Journal of the American Oil Chemists Society,
75(2), 169180.
Sanders, T., Oakley, F., Miller, G., Mitropoulos, K., Crook, D., & Oliver, M. (1997).
Inuence of n-6 versus n-3 polyunsaturated fatty acids in diets low in saturated
fatty acids on plasma lipoproteins and hemostatic factors. Arteriosclerosis,
Thrombosis, and Vascular Biology, 17(12), 34493460.

131

Simopoulos, A. P. (2002). The importance of the ratio of omega-6/omega-3 essential

fatty acids. Biomedicine & Pharmacotherapy, 56(8), 365379.
Wang, G., & Wang, T. (2008). Oxidative stability of egg and soy lecithin as affected
by transition metal ions and pH in emulsion. Journal of Agricultural and Food
Chemistry, 56(23), 1142411431.
Willis, W. M., & Marangoni, A. G. (1999). Assessment of lipase-and chemically
catalyzed lipid modication strategies for the production of structured lipids.
Journal of the American Oil Chemists Society, 76(4), 443450.
Yang, D., Gan, L. J., Shin, J. A., Kim, S., Hong, S. T., Park, S. H., et al. (2013).
Antioxidative activities of Ginkgo biloba extract on oil/water emulsion system
prepared from an enzymatically modied lipid containing alpha-linolenic acid.
Journal of Food Science, 78(1), C43C49.
Yang, D., Shin, J. A., Gan, L. J., Zhu, X. M., Hong, S. T., Sung, C. K., et al. (2012).
Comparison of nutritional compounds in premature green and mature yellow
whole wheat in Korea. Cereal Chemistry, 89(6), 284289.
Zahradnkov, L., Schmidt, ., Sekretr, S., & Janc, L. (2007). Determination of the
antioxidant activity of Ginkgo biloba leaves extract. Journal of Food and Nutrition
Research, 46, 1519.